DI4S-396N 82 100S8-0563802.00 00 NEL Copyright | OSURGIRY 1982 by the Congress of Neurologieal Surgeons Vol. 10. No. 5, 1982 Printed in U.S.A. Ischemic Complications after Combined Internal Carotid Artery Occlusion and Extracranial-Intracranial Anastomosis Fernando G. Diaz, M.D., Ph.D., James I. Ausman, M.D., Ph.D., and Jeffrey E. Pearce, M.D. Department of Neurosurgery, Henry Ford Hospital, Detroit, Michigan Seven of 120 aneurysm patients admitted to the Henry Ford Hospital from October 1978 to August 1981 had giant internal carotid artery aneurysms that were treated by a combined internal carotid artery occlusion and extracranial-intracranial anastomosis. Three of these patients developed postoperative ischemic complications during the progressive closure of the carotid artery. These complications included the transient onset of syncope, hemiparesis, hemisensory deficits, and dysphasia. These complications resolved after the clamp was reopened and/or intravenous heparin was given. The possible mechanisms involved in the development of ischemia included the development of emboli at the occlusion site or inadequate flow originating from the area of the anastomosis. Prolonged occlusion of the vessel over a 7- to 10-day course with concurrent administration of intravenous heparin is recommended. (Neurosurgery 10:563-570, 1982) Key words: Cerebral ischemia, Extracranial anastomosis, Giant aneurysm, Internal carotid artery occlusion, Intravenous heparin, Ischemic complication, Prolonged vessel occlusion In the past, giant intracavernous carotid aneurysms have been either left untreated (11) or treated by carotid ligation (6, 7, 12, 13, 16). These aneurysms have a very low incidence of rupture, and many patients live without additional problems (11), although they may have to accept neurological symptoms such as headache, blindness, and a partial or complete ipsilat- eral ophthalmoplegia (3, 4, 6, 18). Furthermore, although carotid artery ligation has been advocated as a treatment for these aneurysms, it carries a subsequent 14 to 60% risk of cerebral ischemia (1. 5, 6, 12, 13, 16). This paper discusses the ischemic complications observed in three of seven patients with giant internal carotid artery aneu- rysms during combined internal carotid ligation and extracran- ial-intracranial (EC-IC) bypass. CLINICAL MATERIAL Seven of 120 aneurysm patients admitted to Henry Ford Hospital from October 1978 to August 1981 had giant aneu- rysms of the internal carotid artery. A general description of their clinical presentation appears in Table 1. Three of the seven patients developed ischemic complications during com- bined internal carotid artery occlusion and EC-IC anastomosis. A detailed description of these three patients is given below. Case 4 A 47-year-old woman had a long-standing history of head- aches and a |-year history of intermittent right upper extremity paresis, She developed diplopia and right eye pain during the month before admission. but her general physical examination was within normal limits. Although she complained of diplopia on right lateral gaze, the extraocular movements appeared normal. A transfemoral four-vessel angiogram showed a giant right intracavernous aneurysm (Fig. 1). She also had a smaller left supraclinoid internal carotid artery aneurysm. A computed tomographic (CT) scan confirmed the presence of a space- occupying lesion in the right temporal fossa (Fig. 2). After a Selverstone clamp was placed on the right internal carotid artery to reduce the carotid flow by 50% (Fig. 3). a right superficial temporal-middle cerebral artery anastomosis was performed Postoperatively. the patient did well, and the clamp was turned daily. However. before the clamp had been closed completely (when the vessel was probably about 90% oc- cluded), the patient developed syncope and left arm and leg numbness on the 7th postoperative day. The clamp was re- opened to the initial postoperative setting (estimated 50% flow reduction) so that the symptoms could resolve. After she was placed on intravenous heparin and her vascular volume was expanded with plasma, the clamp was gradually and success- fully closed. The postocclusion angiogram showed that the internal carotid artery was completely occluded, the aneurysm was no longer filled, and the superficial temporal to middle cerebral artery bypass was patent (Fig. 4). Since her discharge, the patient has had minor right retro-ocular headaches and the diplopia has persisted. Case 5 A 34-year-old woman developed left retro-ocular and fron- totemporal headaches 3 months before admission. Two weeks later she had a syncopal event preceded by severe headache and was taken to another hospital. A CT scan showed a large space-occupying lesion in the left temporal fossa, which on angiography proved to be a giant aneurysm originating from the supraclinoid portion of the left internal carotid artery (Fig. 5). An attempt at direct surgical obliteration was unsuccessful because the aneurysm was too large. The patient was referred to our hospital for treatment. On admission, she had a mild residual right central facial paresis, but was otherwise neurologically intact. A repeat angiogram confirmed the presence of the aneurysm and the absence of a superficial temporal artery. A Selverstone clamp was placed on the left internal carotid artery, and a saphenous vein autograft was placed between the left external carotid artery and the left middle cerebral artery. The Selverstone clamp was turned daily after the patency of the graft had been proven angiographically. On the Sth postoperative day, the patient developed a tran- sient right sensory deficit (when there was approximately 85% vascular occlusion). which resolved after the clamp was re- opened enough to allow a 50% reduction in blood flow. After she was placed on intravenous heparin and her vascular volume was expanded with plasma. the clamp was gradually and successfully occluded. The postocclusion common carotid an- giogram showed that the internal carotid artery was completely occluded by the clamp, the saphenous vein graft was filling the entire middle cerebral tree. and the aneurysm was no longer 564 DIAZ et al. Neurosurgery, Vol. 10, No. 5 filled (Fig. 6). She has remained asymptomatic since the pro- cedure. ard CN palsy Case 6 NI NI NI NI NI A 79-year-old man developed a series of successive severe left frontal headaches approximately 3 days before admission. Twenty-four hours later, he had a syncopal event associated with a severe headache. He was admitted to another hospital where he developed left facial numbness and left ptosis. On examination he had a right facial palsy, an absent left corneal reflex, bilateral Babinski signs, and a partial left 3rd nerve palsy, which became complete over the next 12 hours. The rest of his neurological examination was negative. A CT scan showed a space-occupying lesion in the left parasellar and temporal areas. A spinal tap was negative, and a cerebral angiogram showed a left giant intracavernous aneurysm (Fig. 7). On admission to our hospital, his neurological examination was normal with the exception of the mild right facial palsy. A Tepeat angiogram showed less filling of the aneurysm, but the mass effect on the CT scan remained unchanged. After a Selverstone clamp was placed on the left internal carotid artery, an anastomosis was performed between the left superficial temporal and the middle cerebral arteries. Over the next few days the clamp was gradually closed. Twelve hours after the clamp had been completely closed, the patient developed a right hemiparesis and dysphasia. He was treated with heparin and volume expansion. An angiogram confirmed that the internal carotid artery was totally obliter- ated, and the superficial temporal artery filled the entire middle cerebral artery (Fig. 8). In the next few days his hemiparesis and dysphasia disappeared, and he was discharged. A follow- up CT scan revealed a small capsular infarct. Three weeks after discharge, he developed a transient right upper extremity paresis. He continues to have a mild right facial palsy, but is otherwise asymptomatic. + a + + Day P/O SS P/O Sx Syncope. 1 + + + ro Patency 4/79 1/29/80 V1/11/80 /8/81 1/30/81 S/8/81 10/20/8! <4 Sa ¢ + 22 Vein- the Internal Carotid Artery* STA MCA Bypass + + eZ “ + Clamp + * zs f + $ + _ normal ¢ CC C DISCUSSION Aneurysm Loo RICY Le R R RC RCC Fi 5 < a 2 & S Eg & 2 2 3 a € 3 g Giant intracavernous aneurysms present a great challenge to the neurosurgeon because of their location and the problems associated with their direct surgical approach. These aneurysms have a tendency to enlarge (3, 7, 9, 11, 12, 18). They can also lead to the development of subarachnoid hemorrhage (3, 12, 13) and are usually associated with palsies of the 3rd, 4th, and 6th cranial nerves (3, 4, 6, 18). In a series of It patients with headaches and cranial nerve palsies, Morley and Barr found that 4 showed no change in neurological deficit on prolonged follow-up (11). Three of these patients improved sponta- neously, 3 became worse, and | patient died. Various surgical approaches have been used to treat these patients. The proximal ligation of the common carotid artery (6, 12, 13, 16) or the internal carotid artery (6, 7, 12, 16) has been tried with variable success. These aneurysms have been trapped (1, 5, 7) by ligating the common or the internal carotid artery in the neck and the internal carotid artery distal to the aneurysm. The aneurysms have been wrapped to prevent fur- ther expansion of the mass (1, 7). A direct attack with ligation of the neck has been performed in a few cases (4). A combined ligation of the internal carotid artery and anastomosis of the superficial temporal to the distal middle cerebral artery has been advocated to treat these aneurysms (1, 6, 7, 18) and in general has been considered safe. The middle cerebral artery anastomosis can be completed using the superficial temporal artery or a saphenous vein graft anastomosed proximally to the common carotid, external carotid, or subclavian artery in the neck (8, 16, 17). Various problems during the course of ligation of the com- ae mon or the internal carotid artery have been noted (10). In the R Temp mass R Temp mass mass L Temp mays L Temp é @ Syncope R neck L 3rd. Sth. R CN R 6th R 3rd. 6th R 7th 7th Diplopia + a re ¥ SAI + jed for 12 hours. HBP. + + + M4 1” Ae So 68 58 Pe 4k Sex Case Fic. 1. Case 4. Preoperative lateral (/eft) and anteroposterior (right) subtraction angiograms show the collection of contrast agent in the right temporal fossa, extending from the cavernous-carotid region to approximately 4/2 cm laterally. This area is equivalent to that seen on the CT scan. Fic. 2. Case 4. Preoperative CT scans. Left, nonenhanced CT scan shows the suggestion of an area of increased density in the right medial temporal fossa. Center, axial CT scan after the administration of contrast medium. A definite area of increased density is demonstrated in the right medial temporal fossa extending from the region of the cavernous sinus laterally. Right, coronal CT scan after the administration of contrast agent. Notice the large mass located in the right temporal region, extending again from the region of the cavernous sinus to approximately 4 cm laterally in the temporal fossa. 200 ™ nin — —200 nin ated eee AAA — 4 Mean Q=75 M/qi, Occluded Mean Q=40 My Fic. 3. Case 4. Actual tracing of the electromagnetic flow measurement recorded from the internal carotid artery during the placement of a Selverstone clamp. Notice the approximately 50% reduction in flow established by first occluding the vessel completely (arrow) and then allowing a gradual resumption of blood flow. Mean Q = blood flow, millilitres per minute. (We used a Cliniflow electromagnetic flow meter, Model 611D (Carolina Medical, King, North Carolina).) 565 566 DIAZ et al. Fic. 4. Case 4, Postoperative an- giogram shows complete oblitera- tion of the internal carotid artery by the Selverstone clamp placed at the origin of the right internal ca- rotid artery. Notice the filling in the middle cerebral artery territory by the EC-IC anastomosis, Neurosurgery, Vol. 10, No. 5 Fic. 5. Case 5. Preoperative subtraction angiogram shows an area of contrast collection originating from the cavernous internal carotid artery. Notice the trilobulated appearance of the aneurysm. The area opacified by the contrast agent is approximately one-third of that observed on the CT scan. report of the Cooperative Study, patients who underwent the gradual occlusion of the internal carotid artery had a 41% incidence of ischemic cerebral complications, whereas those who underwent the gradual ligation of the common carotid artery had only a 24% incidence (12). When the carotid artery was occluded abruptly, the rate of complications for ligation of the internal carotid artery rose to 59% and that for ligation of the common carotid artery rose to 32%. Those patients who had a ruptured aneurysm had a much greater incidence of complications than those who had a nonruptured aneurysm (12). In 96% of the cases there was an 8- to 14-day interval from the time the clamp was placed until occlusion. Ninety- five per cent of the ischemic complications occurred during the first 7 days of progressive closure of the clamp (12). May 1982 ISCHEMIA AFTER ICA OCCLUSION & EC-IC ANASTOMOSIS 567 Fic, 6. Case 5. Postoperative lateral (/eft) and anteroposterior (right) subtraction angiograms show the complete obliteration of the left internal carotid artery by the Selverstone clamp. The middle cerebral artery tree is filled by patent saphenous vein graft. 5/4/81 Fic. 7. Case 6. Preoperative left carotid angiogram shows a large intracavernous aneurysm projecting into the temporal fossa 568 DIAZ et al. Fic. 8. Case 6. Postoperative oblique left carotid angiogram shows complete occlusion of the internal carotid artery, with no angiographic visualization of the aneurysm and excellent filling of the middle cerebral artery territory by a patent bypass. The ischemic complications reported include the develop- ment of transient neurological deficits including contralateral mono- or hemiparesis, central facial palsy, or dysphasia, which generally have been short-lived (5, 6, 12, 16). In some cases the neurological deficits have been permanent (5, 12, 16). The reported frequency of transient ischemic events has been 11% (5) to as high as 20% (16). The rate of permanent deficits has generally been about 12 to 15% (5, 16). Without a doubt, the abrupt occlusion of the common or internal carotid artery can lead to more ischemic complications of any type (12, 13, 15, 16). The management of giant intracranial aneurysms with a combined internal carotid artery ligation and EC-IC anasto- mosis offers some distinct advantages. The ligation of the carotid artery in the neck decreases the flow entering directly into the aneurysm sac, decreases the local pressure, and de- creases the size of the aneurysm as well as its potential for rebleeding (12. 18). It has been observed that ligation of the common carotid artery leads to thrombosis of the aneurysm (14). The drop in flow through the internal carotid artery associated with the development of local thrombosis can cer- tainly obliterate the aneurysm from the circulation and resolve the problem. This combined effect carries with it the potential for extending the thrombus beyond the ostium of the aneurysm and for leading to the development of emboli or progressive extension into the internal carotid artery distal to the aneurysm (9). This effect is generally not seen when the flow that has been reinstituted through the EC-IC anastomosis is sufficient Neurosurgery, Vol. 10, No. 5 to maintain adequate circulation through the area of slowest flow. near the bifurcation of the internal carotid artery. Spetzler et al. reported that none of 13 patients who were treated by Selverstone clamp placement and EC-IC bypass developed ischemic complications after the clamp was abruptly occluded on the 3rd day (18). In 3 of our 7 cases (Table 1), transient ischemic symptoms resolved after the clamp was reopened and gradually reclosed (in 2) or after heparin admin- istration and volume expansion only (in 1). Even with an angiographically patent anastomosis, it takes time for the cere- bral hemodynamics to accommodate to the new route of flow. If sufficient time is allowed for the bypass to expand, the potential for ischemic complications is reduced. As was previ- ously indicated, 95% of the ischemic complications after carotid ligation occur within the first 7 days of the progressive occlu- sion. If sufficient time is allowed for the bypass to expand, in some of our cases the superficial temporal artery will expand 2 to 3 times within 7 days of the anastomosis (2). The anasto- mosis of the saphenous vein to the middle cerebral artery produces a greater amount of flow and pressure. Even in these cases, the potential for ischemic complications exists because the recipient middle cerebral artery branch may also require time to expand and accommodate to this greater amount of flow. In conclusion, we caution that those considering the com- bined Selverstone clamp ligation of the internal carotid artery and EC-IC anastomosis for treatment of giant intracranial aneurysms should occlude the internal carotid artery clamp gradually over 7 to 10 days to decrease the potential develop- ment of ischemic complications. A strong case can be made for starting full heparinization after angiographic patency of the bypass has been established and continuing its prophylactic use until after the clamp has been totally occluded for at least 24 hours. This procedure could then be followed by the admin- istration of antiplatelet agents (which were not used in our seven patients). Received for publication, December 9, 1981; accepted, January 18, 1982. Reprint requests: Fernando G. Diaz, M.D., Ph.D., Henry Ford Hospital, Department of Neurosurgery, 2799 West Grand Boulevard, Detroit, Michigan 48202. REFERENCES 1, Ammerman BJ, Smith DR: Giant fusiform middle cerebral aneu- rysm: Successful treatment utilizing microvascular bypass. Surg Neurol 7:255-257, 1977. 2. Ausman JI, Latchaw RE, Lee MC, Ramirez-Lassepas M: Results of multiple angiographic studies on cerebral revascularization pa- tients, in Schmiedek P (ed): Microsurgery for Stroke. New York, Springer-Verlag, 1977, pp 222-229. 3. Barr HWK, Blackwood W, Meadows SP: Intracavernous carotid aneurysms: A clinical-pathological report. Brain 94:607-622, 1971. 4. Benedetti A, Curri D: Direct attack on carotid ophthalmic and large internal carotid aneurysms. Surg Neurol 8:49-54, 1977. 5. Drake CG: Giant intracranial aneurysms: Experience with surgical treatment in 174 patients. Clin Neurosurg 26:12-95, 1979, 6. Gelber BR, Sundt TM Jr: Treatment of intracavernous and giant carotid aneurysms by combined internal carotid ligation and extra- to intracranial bypass. J Neurosurg 52:1-10, 1980. 7. Little JR, St Louis P, Weinstein M, Dohn DF: Giant fusiform aneurysm of the cerebral arteries. Stroke 12:183-188, 1981. 8. Lougheed WM, Marshall BM, Hunter M, Michel ER, Sandwith- Smyth H: Common carotid to intracranial internal carotid bypass venous graft: Technical note. J Neurosurg 34:114-118, 1971. 9. Mehdorn MH, Chater NL, Townsend JJ, Darroch JD, Perkins RK. Lagger R: Giant aneurysms and cerebral ischemia. Surg Neurol 13:49-57, 1980. May 1982 10. Miller JD, Jawad K, Jennett B: Safety of carotid ligation and its role in the management of intracranial aneurysms. J Neurol Neu- rosurg Psychiatry 40:64-72, 1977. I. Morley TP, Barr HWK: Giant intracranial aneurysms: Diagnosis, course, and management. Clin Neurosurg 16:73-94, 1968. 12. Nishikoa H: Report on the Cooperative Study of Intracranial Aneurysms and Subarachnoid Hemorrhage: Section VIII, Part 1. Results of the treatment of intracranial aneurysms by occlusion of the carotid artery in the neck. J Neurosurg 25:660-682, 1966, 13. Odom GL, Tindall GT: Carotid ligation in the treatment of certain intracranial aneurysms. Clin Neurosurg 15:101-116, 1967. 14. Pozzati E, Fagioli L, Servadei F, Gaist G: Effect of common carotid ligation on giant aneurysms of the internal carotid artery: Computerized tomographic study. J Neurosurg 55:527-531, 1981. 15. Pritz MB: Cardiopulmonary monitoring during graded cervical internal carotid artery occlusion: Physiological results and thera- peutic implications. Neurosurgery 8:520-524, 1981. 16. Roski RA, Spetzler RF, Nulsen FE: Late complications of carotid ligation in the treatment of intracranial aneurysms. J Neurosurg 54:583-587, 1981. 17, Spetzler RF, Rhodes RS, Roski RA, Likavec MJ: Subclavian to middle cerebral artery saphenous vein bypass graft. J Neurosurg 53:465-469, 1980. 18. Spetzler RF, Schuster H, Roski RA: Elective extracranial-intracra- nial arterial bypass in the treatment of inoperable giant aneurysms of the internal carotid artery. J Neurosurg 53:22-27, 1980. COMMENTS The importance of recognizing that an EC-IC bypass may not be capable of preventing ischemic symptoms when a nor- mal major feeding artery is occluded cannot be overemphasized (1). We reported 21 cases of EC-IC bypass as part of the elective treatment for giant intracranial aneurysms (3). In none of these patients did we encounter immediate or delayed ischemic complications. Shortly after the paper was in press, we did Case 22. This patient presented with a 3rd nerve palsy and severe orbital pain from a large cavernous sinus aneurysm. Because of time constraints, we shortcut our own protocol and occluded the internal carotid artery (ICA) 2 days after the EC- IC bypass. Thirty-six hours later the patient developed a sud- den contralateral hemiplegia. Fluid push and heparin were of no appreciable help and the patient was left with a significant permanent neurological deficit. Since then we have had two more cases in which we followed our usual routine without further ischemic complications. In our original publication we recommended that the ICA be sufficiently occluded to either lower blood flow by half as measured with a flowmeter or decrease the stump pressure to halfway between the open pressure and the completely oc- cluded pressure (2). Either method requires more than 85% closure of the cross sectional area of the ICA. This stump effect enhances the development of the EC-IC bypass. Three or four days after this maneuver, a repeat angiogram is performed and, if the EC-IC bypass is patent, then the [CA is completely occluded. The authors’ suggestion of gradually closing the clamp is very reasonable. The problem is that there is very little room between 90% and 100% closure, making gradual occlusion unreliable. Their treatment of volume expansion and heparin administration is judicious. It is important obviously to keep the patient well hydrated after the EC-IC bypass. We routinely start anti-platelet aggregation medication | to 2 weeks before operation. A most important point in their paper that should not be overlooked in the discussion of the ischemic problems is that none of these patients had any permanent neurological deficits secondary to ICA occlusion. Further experience is required ISCHEMIA AFTER ICA OCCLUSION & EC-IC ANASTOMOSIS 569 before a consensus can be reached on what is indeed the most favorable interval between the performance of an EC-IC by- pass and the safe occlusion of an ICA. The rationale, however, for performing an EC-IC bypass before ICA occlusion is sound and is supported by their paper. Robert F. Spetzler, M.D. Cleveland, Ohio 1, Samson DS, Neuwelt EA, Beyer CW, Ditmore QM: Failure of extracranial-intracranial arterial bypass in acute middle cerebral artery occlusion: Case report. Neurosurgery 6:185-188, 1980. 2. Spetzler RF, Schuster H, Roski RA: Elective extracranial-intracra- nial arterial bypass in the treatment of inoperable giant internal carotid artery aneurysm. J Neurosurg 53:22-27, 1980. 3, Spetzler RF. Schuster H, Takaoka Y: The role of EC-IC in the treatment of giant intracranial aneurysms. Neurol Res 2:345-359, 1980. This is a provocative paper, raising questions in several areas of ongoing controversy but, unfortunately, not providing cer- tain answers. The questions surgeons face when contemplating carotid ligation in the treatment of giant aneurysms are: (a) the relative importance of hemodynamic and microembolic causes of cere- bral ischemia after carotid ligation; (b) the value of prophylactic extracranial to intracranial arterial bypass in preventing cere- bral ischemia; (c) the predictability of risk with intentional carotid occlusion; and (d) the value of prophylactic anticoag- ulation, platelet suppression, and volume expansion in the prevention and treatment of cerebral ischemic events following carotid occlusion. Our experience at the University Hospital in London is at variance with that of the authors. Of 49 giant carotid-ophthal- mic aneurysms, 35 could be directly attacked and the neck clipped. Fourteen patients underwent treatment using the prin- ciple of Hunterian ligation by occlusion of the internal carotid artery either by a clamp in the neck or, more recently, with balloon occlusion at the neck of the aneurysm. We have found it necessary to do prophylactic EC/IC bypasses in only 3 of these 14 patients. There have been two ischemic complications in the 14 cases, both of which were clearly documented as embolic. The ischemic event was transient in | of these patients and, interestingly, this patient had not had a prophylactic bypass. In the other case, prophylactic bypass had been done and, despite excellent filling of all of the middle cerebral artery tree, the patient suffered a severe permanent neurological deficit when a large embolus lodged in the M-1 segment, occluding vital thalamostriate perforators and causing a deep central infarction. Our experience with giant carotid-cavernous aneurysms is smaller, but [ think points out an important difference between aneurysms below the ophthalmic collateral and those above. We have occluded the carotid artery in 13 patients with carotid- cavernous aneurysms, none of whom have had a prophylactic bypass. There was one transient ischemic complication, again almost certainly due to emboli arising in the distal stump or from the partially thrombosed sac. It is commonly accepted that about one patient in five is at risk for cerebral ischemia when the internal carotid artery is intentionally occluded. In only a fraction of this 20% is the ischemia due only to impaired flow. I believe that the remain- der most commonly are due to thromboembolism from the distal stump of the artery or from the sac of the aneurysm itself. The EC/IC bypass is at least theoretically less attractive in preventing or minimizing thromboembolic ischemia. 570 DIAZ et al. One could argue that carotid occlusion is always dangerous. in both the short and the long term. and should be preceded by a bypass to improve the patient’s chances of avoiding an ischemic event. We. however, believe that the challenge is not to bypass every patient but to select for bypass those at risk and avoid the small but undeniable hazards of a second and possibly unnecessary surgical procedure. The use of cerebral blood flow measurement under tightly controlled and standardized physiological conditions allows patients suitable for permanent carotid ligation to be identified with some certainty. We have used the criteria of the Glasgow group (1). as follows: (a) carotid ligation is always safe if the CBF is greater than 40 ml/minute/100 g during a trial occlu- sion of the carotid artery: (b) the ligation is safe if the CBF during clamping is 20 to 40 ml/minute/100 g provided that the CBF reduction from a control flow is less than 25%; (c) ligation is also safe in the CBF range of 20 to 40 ml/minute/100 g with up to a 35% reduction in flow provided that the internal carotid artery pressure is greater than 60 mm Hg; and (d) ligation is always unsafe if the CBF during clamping is less than 20 ml/ minute/100 g regardless of the change from the control flow. These criteria have proved to be useful and reliable in our hands, but will not predict those patients who develop ischemic complication from thromboembolism. Studies in our Department have demonstrated active plate- let-fibrin deposition and turnover in the sacs of giant aneurysms and the potential for this material to embolize into the cerebral Neurosurgery, Vol. 10, No. 5 vascular tree (2). For this reason. our practice has been to trap these aneurysms whenever possible or to occlude the parent artery as close to or. preferably. at the neck of the aneurysm with a detachable balloon to prevent the egress of embolic material and to minimize the dead space in the stump of the remaining parent vessel. As well. we think that it is eminently reasonable to pretreat these patients with platelet antiaggregant drugs and heparin as soon as the general condition of the wound permits. Finally, we have not considered it important to gradually occlude the carotid artery or to wait for some hypothetical maturation of the bypass with time. Rather, the proximal vessel is occluded abruptly as soon as the graft is shown to be patent and usually within 24 hours of the micro- vascular anastomosis. It would seem that nothing is as effective in producing good flow across the graft as a significant pressure drop, and this is ensured by immediate proximal vessel occlu- sion. S. J. Peerless, M.D. London, Ontario 1, Miller JD, Jawad K, Jennett B: Safety of carotid ligation and its role in the management of intracranial aneurysms. J Neurol Neu- rosurg Psychiatry 40:64-72, 1977, . Sutherland GR, King M, Vezina W, Peerless SJ, Brown G, Cham- berlain MJ: Platelet interaction within giant intracranial aneu- rysms. J Neurosurg (in press). N